1.4 Exact ODES.

Integrating Factors
Exact ODE:

Definition: A first order ordinary differential equation of the form

M (x, y) dx + N (x, y) dy = 0 ……………………..(1)

is called exact if there exists a function u(x, y) such that

du(x,y) = M (x, y) dx + N (x, y) dy ……………..(2)

Where du is called the differential of u(x,y)

Since,

u u
du  dx  dy
x y ………………….….(3)

where u = u (x, y)

Then from equations (1) and (2)

du = 0

Integrating, u = c (constant)

i.e. u (x, y) = c, which is the general solution of equation (1).

Example -1:-

xdy + ydx = 0

The LHS of the given equation is d (xy)

So, d (xy) =0

By integration, xy = c (constant)

which is the general solution of the given differential equation

Also, the given equation is an exact ODE.

Necessary and Sufficient condition for exactness:

Let us suppose equation (1) is exact, then

du(x, y) = M (x, y) dx + N (x, y) dy (by definition)

1
 u u
 dx  dy  M ( x, y )dx  N ( x, y )dy
x y

Comparing both sides of the equation,

u u
  M ( x, y ) and  N  x, y 
x y .……………….(4)

Differentiating M(x,y) w.r.to y and N(x, y) w. r. to x we have

M  2u N  2u
 and 
y yx x xy

 2u  2u
Since 
yx xy

M N
 
y x ……………………..(5)

which is not only the necessary but also the sufficient condition for equation (1) to be exact.

Hence, the necessary and sufficient conditions for the differential equation (1) to be exact is

M N

y x

Method of finding u(x,y):–

If the ODE (1) is exact then from equation (4),

u
 M , M  M ( x, y )
x
Integrating both sides w.r. to x, we have

u   Mdx  K ( y )
………………………...(6)
u dK
Differentiating (6) w.r. to y and then comparing with N
, we will get and then K(y) after
y dy
integration.

Similarly, we can also use

u
 N , N  N  x, y 
y

2
Integrating both sides w.r.to y,

u   Ndy  l (x)
…………………….(7)

Where l(x) is the integration constant as x is constant.

u
Differentiating equation (7) w.r.to x and the comparing with  M,
x
dl
we will get and then l(x) after integration.
dx
Reduction to Exact Form : Integrating Factors
Integrating Factor (I.F.)

When the left-hand side of the equation

M (x, y) dx + N(x, y) dy = 0

is not an exact or a total differential, then it is necessary to choose a function µ (x, y) such that after
multiplying by it, the left-hand side of the above equation becomes an exact differential du and is
given by

du = µ (x,y) M (x,y) dx + µ(x,y) N(x,y) dy

Such a function µ (x,y) is called an I.F.

Definition:

An integrating factor is a function when multiplied by it, the left hand side of equation (1) becomes an
exact differential.

Example – 2 :

-ydx + xdy = 0 ……… (1)

Here, M = -y, and N = x

M N
  1 and 1
y x

Since M N

y x
So, the given equation is not exact.

In order to make the equation exact, let us multiply the given equation (1) by 1/x2

 ydx  xdy 1
 2
 0. 2
x x

3
 y 1
 2
dx  dy  0
x x

 y y
 d    0   C (constant)
x x

which is the general solution of the given problem.

Here, µ(x,y) = 1/x2 is the I.F and it is obtained by inspection.

How to Find an Integrating Factor: -

Rule-1:

M N

y x
 f (x)
N
If, a function of x alone,

I .F .  e 
f ( x ) dx
Then

Rule – 2 :

M N

y x
If ,  F ( y)
M
  F ( y ) dy
a function of y alone then I .F .  e

Problem Set 1.4

Test for exactness. If exact, solve If not, use an integrating factor (I.F) to solve.

Q-1 2xy dx + x2 dy = 0

Solution:-

2xy dx + x2 dy = 0…………….. (1)

 
 d x2 y  0

Integrating, we have

x2y = C is the general solution.

4
 Or

The given equation is of the form M (x, y) dx + N (x, y) dy = 0

So, M = 2xy, N = x2

M N
  2 x and  2x
y x
M N
Since, 
y x

Hence the given equation is exact.

u
By using formula (4), M
y
 u   Mdx  k ( y )
.………………….(2)

 u   2 xydx  k ( y )

 u  x2 y  k  y

Differentiating both sides w.r.to y,

u dk
 x2 
y dy
………………(*)

But
u
 N  x2
y
Substituting in equation (*)

dk
x2  x2 
dy

dk
 0
dy

 k  c (constant )

So, from equation (2)

u = x2y=c is the general solution

2
i.e. y  c x

5
Q – 2: x3dx + y3dy = 0

Solution: -

Comparing with the equation M (x, y) dx + N (x, y) dy = 0

We have

M = x3 and N = y3

M N
 0 0
y x

Since, M  N
y x

So, the given equation is exact

Since,

u  Mdx  K  y 

i.e u  x 3dx  K  y 

x4
u  K  y
4

Differentiating both sides w.r.to y,

u dk  K dK 
  y  dy 
y dy  
u
But  N  y3
y

dK
So  y3
dy
Integrating w. r.to y,

dK
 dy dy   y dy
3

y4
K C
4

x 4 y4
u  C
4 4
6
is the general solution.

Q-5 (x2+y2)dx – 2xydy = 0 ….(1)

Solution:

Here, M = x2 + y2, N = - 2xy

M N
 2y,  2y
y x

M N
Since, 
y x
So, the equation is not exact.

M N

y x 2y  2y 4y 2
     f x
N 2xy 2xy x
Now

2
 f  x  dx   dx
So, I.F  e  e x  e 2ln x
2
 eln x

 1 x2

Now multiplying 1/x2in equation (1),

1 2 2xy
x 2 x

x  y 2 dx  2 dy  0

 y2  2y
  1  2  dx  dy  0
 x  x
…(2)

Which is of the form M dx + Ndy = 0

Now equation (2) is exact.

So ,
 y2 
u  1  2  dx  K  y 

 x 

7
   y2 
 x   K  y
 x 

y2
x  K  y
x

Now differentiating both sides w. r. to y

u  2 y K
  N
y x y

K
 0KC
y
Hence, the general solution is

y2
u  x, y   x  C
x

Where c is a constant.

*******************************END*******************************************